The quantification of learning horrifies this statistician. The belief that a standardized test is more objective than say a trained human assessor neglects the subjective nature of human makers of the test. It puts the humans behind a facade that creates the appearance of objectivity but not actual objectivity. The benefit of standardized tests is that they lower labor costs of assessment not that they increase the quality of assessment. It's only with that admission that a reasonable cost-benefit analysis could be done.
Your argument is that standardized tests are more biased than individual assessment? I find that hard to believe bordering on ridiculous. There's a great deal of evidence that standardized exams are less biased as there's a consistent standard of correct answer across the board.
With individual assessment there's far too much room for "I know what you meant" to sneak in. This has come up a lot with job interviews; if you want to interview impartialy, have a fixed bank of questions and grade answers to a rubric.
The argument is that an exam with free form answers is a more objective measure of knowledge than a multiple choice exam. Not an argument about bias.
I think it depends on the subject matter, especially interpretation of art is hard to have one correct answer, an essay test could be graded upon support for the argument etc, where a multiple choice test is really "what did the authors of the test think the author intended?"
Agreed. Unless we're talking about math (logic proofs), we should probably stop using the term, "objective." It's sloppy thinking.
Because standardized tests can be applied to such a large set, the biases of one test maker can be spread to many test takers. Contrast this with individual assessment where the biases of many test "makers" have an opportunity to cancel each other out. This is similar to the logic behind randomized assignment in randomized-control experiments. It's not that the mice are the same, it's that asymptotically their differences cancel each other out. I'd recommend this article: http://stat.wharton.upenn.edu/~rosenbap/heteroReprint.pdf